For many companies, the last decade of application evolution has focused on the use of horizontal tools and software to build applications. Service-oriented architecture, enterprise service bus, application lifecycle management and unified communications are examples of this trend. The problem is that a horizontal focus tends to make every business application a massive integration function, crossing application and vendor lines. Smart process applications take a different approach, focusing on the vertical and agility within an application's boundaries. Do the following to get the most from application modernization based on smart process applications:
- Know when you have a smart process
- Identify your dominant business process variables
- Beware of horizontal anarchy
Business process support through applications has always had to balance horizontal and vertical issues. While most business processes create what's almost a specialist ecosystem within the workers involved, implementing one-off solutions for every business activity has never been seen as efficient.
Architects have addressed business process support through mechanisms such as componentization of applications and orchestrations of workflow (e.g., Business Process Execution Language). Smart process applications take this to the next level, focusing not on adapting horizontal applications to vertical needs, but on the fulfillment of vertical business process needs as the paramount goal.
Smart process applications typically assemble all of the support needed for a business process into a single tool or application, including the following:
- Core elements such as customer relationship management and manufacturing resource planning
- A collaborative framework to coordinate organizational activity around the core
- Analytics processes to drive decisions and manage the activities
These three things are essential to smart process applications, but because they're also crucial to supporting any business process, getting smart process applications right means digging deeper into each possible mission.
Candidates for smart process application implementation should also be assessed for cloud implementation.
Targeting the right processes
Enterprise architecture can define candidates by selecting processes with an exchange of work internal to a group of employees and linked to the rest of the company's activities via selected exchanges. A smart process candidate is one that can be considered somewhat autonomous from a company workflow perspective.
Some companies have pointed out that the cloud can be a useful guide for finding smart process application candidates. Most applications that can be readily supported in Software as a Service form (CRM, for example) will likely be suitable for smart process application implementation as well. Thus, candidates for smart process application implementation should also be assessed for cloud implementation.
The next step in a smart process application project is to identify the dominant business process variables the application will support. Variables, in this scenario, means the broader context of the term -- the most dynamic things about the business processes being targeted.
All smart process applications wrap application-specific workflows in contextual business activities (such as unified communications and collaboration, social networking, etc.), and the most important and effective wrappers will be those targeting significant business dynamics. For example, if your process candidate involves a significant amount of review and coordination, then your smart process application should be built around collaborative tools. If it's dominantly about data mining, then select tools built around analytics capability.
A review of the business process variables will help determine how much the activities of a candidate business process spread beyond business-unit boundaries. Remember, a smart process application is designed to support a process and not an entire business. How this application interacts with other business processes will determine how much utility can be gained through its use in the first place.
Making the case for smart process applications
One of the primary arguments in favor of smart process applications is the ease with which they adapt to changes in the business process. If the adaptation of one business process changes a wide range of adjacent processes, then architects will have to consider this broad impact and adapt the applications supporting the adjacent processes. This quickly kills any claim of agility.
If a review of business variables determines there is considerable overlap between a candidate process and adjacent activities, it may be necessary to broaden the scope of the smart process application or reconsider how the candidate processes modernization.
The notion of "impact creep" from a smart process application into adjacent business processes is an example of the dualistic nature of application modernization. Do you componentize and compose horizontal tools to achieve vertical process efficiency? Should you harmonize vertical solutions to generate horizontal harmony across business processes?
The more smart process applications you define, the greater the risk you'll end up with vertical silos of information that inhibit broad business management effectiveness. This is particularly true of collaboration and analytics implementations in smart process applications, because both activities are naturally broad in their connection to the rest of the business.
Handling horizontal integration
One way to avoid horizontal anarchy is to limit the number of smart process applications being used and focus on applications that are relatively isolated. Most companies will find that only two or three business processes link effectively to smart process applications, while the rest spread activity too far across business units. Focus on the best candidates, which are those with the most internal work exchange and the most contained external exchange.
Another positive step in managing horizontal integration is to look for smart process applications that integrate with the critical collaboration and analytics tools in a general way, rather than providing a customized strategy for either or both. That will ensure that information exchanges and data collection can be unified across all smart process applications and the other application tools used to support the business. This becomes more important as the number and scope of smart process applications increase, so it should be a part of your overall modernization strategy to watch horizontal integration carefully.
There are no perfect solutions to application modernization problems, and smart process applications are no exception. When they do work well, though, they can jump-start an application modernization process by bringing agility and integration to the most critical parts of your business.
About the author:
Tom Nolle is president of CIMI Corp., a strategic consulting firm specializing in telecommunications and data communications since 1982.